Thesis – Household Recycling Behaviour and Plastic Waste in Australia

Abstract

Plastic packaging waste is a persistent problem for Australian kerbside recycling systems, where contamination undermines the quality of recovered material and limits progress toward national resource recovery targets. This study examined the determinants of household recycling behaviour across metropolitan and regional Victoria, combining a survey of 512 households with a physical audit of the kerbside recycling bins of 120 of those households. The Theory of Planned Behaviour, extended with recycling knowledge and dwelling type, framed the analysis. Perceived behavioural control and recycling knowledge were the strongest predictors of self-reported behaviour, and households in apartments recycled less than those in separate houses. The waste audit recorded a contamination rate of 13.2 per cent, exceeding the level at which reprocessors down-grade or reject loads. The findings indicate that convenience-oriented nudges, clearer labelling, targeted education and container deposit incentives are likely to reduce contamination more effectively than attitude-focused campaigns alone.

Introduction

Australia generates a large and growing volume of plastic waste, yet recovers only a small share of it. National reporting estimates that the country produces roughly 2.6 million tonnes of plastic each year and recycles well under a fifth of it, a recovery rate far below that achieved for paper, metals and glass (Australian Bureau of Statistics [ABS], 2022; Department of Climate Change, Energy, the Environment and Water [DCCEEW], 2022). Packaging is the largest single fraction of this stream, reaching households as beverage containers, films and rigid containers that are notionally recyclable but frequently misdirected. Closing this gap is central to the National Waste Policy Action Plan target of an 80 per cent average resource recovery rate across all streams by 2030 (Commonwealth of Australia, 2019), to the National Plastics Plan’s phase-out of problematic single-use plastics (Commonwealth of Australia, 2021), and to the packaging targets pursued by the Australian Packaging Covenant Organisation (APCO, 2023).

Kerbside recycling is the main mechanism through which Australian households participate, and its effectiveness turns on what residents place in the yellow-lidded bin. Contamination, the presence of non-target material such as soft plastics, food waste and general refuse, is the principal threat to kerbside performance. Contaminated loads reduce the value of recovered commodities and, above a threshold of roughly 10 per cent, may be down-graded or rejected by material recovery facilities (Sustainability Victoria, 2023). The problem has been compounded by disruptions to recycling infrastructure, notably the collapse of the REDcycle soft-plastics program in November 2022, which removed a familiar disposal pathway. State responses, including the standardised kerbside streams and container deposit scheme introduced under Victoria’s Circular Economy (Waste Reduction and Recycling) Act 2021, reflect a policy consensus that household behaviour, not merely infrastructure, must change.

Understanding why households recycle well or poorly is therefore a practical question. This thesis examined the determinants of household recycling behaviour in an Australian setting, testing whether a behavioural framework extended with knowledge and situational factors could explain both self-reported behaviour and independently observed contamination. Three research questions guided the study:

1. What levels of recycling attitude, subjective norm, perceived behavioural control, knowledge and behaviour do Australian households report, and how do these differ between separate houses and apartments?
2. Which of these determinants predict self-reported recycling behaviour after accounting for household characteristics?
3. Does independently audited kerbside contamination correspond to the behavioural determinants measured in the survey?

Literature Review

The Theory of Planned Behaviour

The Theory of Planned Behaviour (TPB) remains the dominant psychological account of pro-environmental action (Ajzen, 1991). It holds that behaviour is driven proximally by intention, which is shaped by three antecedents: attitude, subjective norm (perceived social pressure), and perceived behavioural control (PBC), the sense that the behaviour is easy and within one’s capacity. PBC influences behaviour both indirectly through intention and directly, because control reflects real constraints as well as perceptions of them (Ajzen, 2020). Applied to recycling, the framework has consistently identified PBC as a leading predictor, reflecting the situational character of the behaviour: residents recycle correctly when it is convenient and when they know what belongs where (Tonglet et al., 2004). A meta-analysis confirmed that perceived control and situational conditions outweigh general environmental attitudes in explaining actual behaviour (Geiger et al., 2019), which motivates an extended model treating recycling knowledge, an operational form of control, as a distinct determinant alongside the core TPB constructs (Barr, 2007).

Nudges and choice architecture

A complementary literature argues that behaviour can be shifted by redesigning the choice environment rather than by changing underlying attitudes. Nudges, low-cost changes to the way options are presented that steer behaviour without removing freedom of choice, have proven effective in domains from retirement saving to energy use (Thaler & Sunstein, 2008). In the recycling context, this insight favours interventions that make the correct action the easy and default action: standardised bin colours and lids, unambiguous on-pack labelling, and clear point-of-disposal signage. Field experiments show that providing households with normative feedback, comparing their behaviour with that of neighbours, can raise participation and reduce contamination because it engages subjective norms at the moment of action (Schultz, 1999).

Incentives and container deposit schemes

Financial incentives represent a third lever: even modest monetary rewards can shift disposal behaviour by attaching a tangible value to material that would otherwise be discarded (Iyer & Kashyap, 2007). Container deposit schemes (CDS) apply this principle at scale. A refundable deposit, typically ten cents per eligible container, is levied at purchase and returned on redemption, creating an incentive to divert containers from litter and the kerbside bin toward a dedicated, cleaner recovery stream. Schemes now operate in most Australian jurisdictions, including New South Wales (Return and Earn), Queensland (Containers for Change), the long-established South Australian scheme, and Victoria’s CDS Vic, introduced in 2023. The Australian evidence base on how these overlapping systems jointly shape household behaviour remains thin, and few studies have paired attitudinal survey data with independent physical measurement of what households place at the kerb. The present study addresses that gap.

Methodology

Design and sample

A cross-sectional, mixed-methods design combined a household survey with a physical waste audit. The survey sample comprised 512 households recruited across metropolitan and regional local government areas in Victoria during one autumn collection season, stratified to include both high-density and detached housing. Of the responding households, 320 (62.5%) occupied a separate house and 192 (37.5%) occupied an apartment, unit or townhouse, and the mean household size was 2.6 persons. Respondents were the adult who reported primary responsibility for waste disposal, since kerbside behaviour is typically enacted by one household member.

Measures

The survey operationalised the extended TPB. Attitude toward recycling, subjective norm, perceived behavioural control and intention were each measured with multi-item seven-point scales, higher scores indicating a more favourable standing on each construct. Recycling knowledge was assessed with a ten-item sorting task in which respondents classified common household items as accepted or not accepted in the kerbside bin, scored from 0 to 10. Self-reported behaviour was captured as a composite index from 0 to 100, combining frequency of recycling with the reported correctness of sorting. Internal consistency was acceptable to high (Cronbach’s alpha between .78 and .89), and recorded household characteristics included dwelling type, household size and age. Figure 1 presents the conceptual model that guided the analysis, in which attitude, subjective norm and perceived behavioural control predict intention, intention and perceived control predict behaviour, and recycling knowledge enters as an additional determinant.

Waste audit

To provide an objective measure independent of self-report, the kerbside recycling bins of a randomly selected subsample of 120 households were audited over two consecutive collection cycles. On each occasion the set-out was weighed, then manually sorted into target material (accepted paper, cardboard, rigid plastics, glass and metals) and non-target material (soft plastics, food and garden organics, textiles, glass fines and general refuse), with auditors blind to the households’ survey responses. The contamination rate was defined as the mass of non-target material as a proportion of total audited mass.

Analytic strategy and ethics

Analysis proceeded in three stages. Descriptive statistics compared houses with apartments; an ordinary least squares multiple regression tested which determinants predicted the self-reported behaviour index, with standardised coefficients reported and variance inflation factors inspected to confirm acceptable multicollinearity among the correlated TPB constructs; and audited contamination was correlated with the survey determinants to assess convergence between reported and observed behaviour. The study was approved by the administering university’s Human Research Ethics Committee and complied with the National Statement on Ethical Conduct in Human Research: participation was voluntary and informed, responses were anonymous, and audit data were linked to households only through a de-identified code.

Results

Descriptive statistics

Table 1 reports the sample characteristics and key measures by dwelling type. Attitudes were uniformly high, with little difference between houses and apartments, yet this favourable disposition did not translate evenly into behaviour. Perceived behavioural control, knowledge and the behaviour index were all lower among apartment households, consistent with the situational barriers of shared bins, limited storage and reduced individual accountability in higher-density dwellings. The gap in the behaviour index, roughly nine points on the 0 to 100 scale, is the clearest expression of this divergence.

Note. Higher scores indicate more favourable attitudes, stronger norms, greater perceived control, more knowledge, stronger intention and more frequent, more accurate recycling.

Waste audit and contamination

The audit processed 4,180 kg of kerbside recyclables from the 120 audited households across two collection cycles. This corresponds to an average set-out of 4,180 / 240 = 17.4 kg per household per collection. Of the total audited mass, 552 kg was non-target material. The contamination rate was calculated as the mass of non-target material divided by the total audited mass:

Contamination rate = non-target mass / total mass = 552 / 4,180 = 0.132 = 13.2%.

At 13.2 per cent, measured contamination exceeded the level of approximately 10 per cent at which reprocessors commonly down-grade or reject loads (Sustainability Victoria, 2023). Soft plastics were the single largest contaminant by mass, consistent with the loss of a dedicated soft-plastics pathway after the 2022 collapse of REDcycle. Audited contamination corresponded to the survey measures: households with higher perceived behavioural control recorded lower contamination (r = -.34, p < .001), as did those with greater knowledge (r = -.29, p < .001). This convergence between an attitudinal instrument and an independent physical measure indicates that the behaviour index reflected genuine differences at the kerb rather than social desirability alone.

Determinants of recycling behaviour

The regression model predicting the behaviour index was significant and explained close to half of its variance, R² = .48, adjusted R² = .47, F(7, 504) = 66.5, p < .001, with all variance inflation factors below 2.5. As Table 2 shows, perceived behavioural control was the strongest determinant, followed by knowledge and intention, while attitude and subjective norm made smaller though still significant contributions and household size did not predict behaviour. Apartment residence was associated with lower behaviour independent of the psychological determinants, a situational penalty that attitudes alone did not overcome.

Note. N = 512. Reference category for dwelling type is separate house. B = unstandardised coefficient; β = standardised coefficient. Model intercept = 8.5.

Taken together, the results answer the three research questions: attitudes were high but insufficient, the determinants that mattered most reflected capability and convenience, and the audit corroborated the survey by linking observed contamination to the same control and knowledge variables that predicted behaviour.

Discussion

The central finding is that Australian households do not fail to recycle for want of goodwill. Attitudes were strongly positive across housing types, yet perceived behavioural control and knowledge, not attitude, distinguished good recyclers from poor ones, and a contamination rate of 13.2 per cent confirmed that intention frequently does not survive contact with the kerbside bin. This pattern, in which control and situational factors dominate general attitudes, aligns with the international recycling literature (Geiger et al., 2019; Tonglet et al., 2004) and bears directly on Australian policy pursuing the National Waste Policy Action Plan targets (Commonwealth of Australia, 2019).

The first implication favours nudges and choice architecture over persuasion. Because behaviour is constrained more by convenience than by conviction, the most efficient interventions make correct sorting the effortless default (Thaler & Sunstein, 2008). Standardised bin lids and colours, consistent on-pack instructions such as the Australasian Recycling Label, and simple point-of-disposal signage lower the control barrier that this study found decisive (APCO, 2023). Normative feedback comparing households with their neighbours is a further low-cost lever that engages subjective norms at disposal (Schultz, 1999). The prominence of soft plastics among audited contaminants, following the withdrawal of the REDcycle pathway, shows how quickly the absence of a clear route degrades kerbside quality.

The second implication concerns container deposit schemes. The refund that underpins schemes such as Return and Earn, Containers for Change and Victoria’s CDS Vic attaches a concrete value to disposal, converting a diffuse intention into an immediate reason to act (Iyer & Kashyap, 2007). Diverting clean, high-value containers into a dedicated stream reduces the burden on the yellow bin, but incentives alone do not address the knowledge deficits driving contamination of the remaining material. Deposit schemes are best understood as complementary to, not a substitute for, measures that improve sorting of the broader kerbside stream.

The third implication concerns education, targeted where it matters most. Knowledge was among the strongest predictors of behaviour and was independently associated with lower audited contamination, yet it was lower in the apartment households that recycled least. Higher-density housing therefore emerges as a priority setting, where shared bins, transient tenancies and constrained storage depress both control and knowledge. Interventions tailored to apartments, such as body-corporate engagement, in-building signage and induction materials for new tenants, address a structural gap that generic campaigns have not closed.

Several limitations qualify these conclusions. The cross-sectional design precludes causal inference, and the behaviour index, though corroborated by the audit, remained partly self-reported and thus vulnerable to social desirability. The audit covered a subsample in a single season. The study was also confined to one Australian state, and jurisdictional differences in scheme maturity and bin systems may limit generalisability. Longitudinal, multi-state designs that pair intervention trials with repeated audits would strengthen both causal claims and external validity.

Conclusion

This study set out to explain why Australian households recycle as they do, and to test whether a behavioural framework could account for both what residents report and what they place at the kerb. Positive attitudes were near-universal but did not, by themselves, produce clean recycling. Instead, perceived behavioural control, knowledge and intention predicted self-reported behaviour, apartment households recycled less than those in houses, and an independent audit recorded a contamination rate of 13.2 per cent that tracked the same control and knowledge variables identified in the survey. Household recycling is thus a problem of capability and convenience more than of conviction. Policies that make the correct action the easy, clearly labelled default, that extend container deposit incentives, and that target education at higher-density housing are more likely to lift the quality of Australia’s kerbside stream, and to advance national recovery and packaging targets, than campaigns that seek only to strengthen an already favourable attitude. Future work should test these levers directly and measure their effect not in stated intention but in the weighed contents of the bin.

References

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.

Ajzen, I. (2020). The theory of planned behavior: Frequently asked questions. Human Behavior and Emerging Technologies, 2(4), 314-324.

Australian Bureau of Statistics. (2022). Waste account, Australia, experimental estimates. ABS.

Australian Packaging Covenant Organisation. (2023). Australian packaging consumption and recovery data 2020-21. APCO.

Barr, S. (2007). Factors influencing environmental attitudes and behaviors: A U.K. case study of household waste management. Environment and Behavior, 39(4), 435-473.

Commonwealth of Australia. (2019). National waste policy action plan. Department of the Environment and Energy.

Commonwealth of Australia. (2021). National Plastics Plan 2021. Department of Agriculture, Water and the Environment.

Department of Climate Change, Energy, the Environment and Water. (2022). National waste report 2022. DCCEEW.

Geiger, J. L., Steg, L., van der Werff, E., & Ünal, A. B. (2019). A meta-analysis of factors related to recycling. Journal of Environmental Psychology, 64, 78-97.

Iyer, E. S., & Kashyap, R. K. (2007). Consumer recycling: Role of incentives, information, and social class. Journal of Consumer Behaviour, 6(1), 32-47.

Schultz, P. W. (1999). Changing behavior with normative feedback interventions: A field experiment on curbside recycling. Basic and Applied Social Psychology, 21(1), 25-36.

Sustainability Victoria. (2023). Victorian recycling industry annual report 2021-22. Sustainability Victoria.

Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving decisions about health, wealth, and happiness. Yale University Press.

Tonglet, M., Phillips, P. S., & Read, A. D. (2004). Using the theory of planned behaviour to investigate the determinants of recycling behaviour: A case study from Brixworth, UK. Resources, Conservation and Recycling, 41(3), 191-214.